G e n es E n cod i n g G I y ci n e - Ri c h A rabidopsis thaliana

Plant Physiol. (1994) 104: 1015-1025

G e n es E n cod i n g G I y ci n e - Ri c h A rabidopsis thaliana Proteins with RNA-Binding Motifs Are lnfluenced by

Cold Treatment and an Endogenous Circadian Rhythm'

Clifford D. Carpenter, Joel A. Kreps, and Anne E. Simon*

Department of Biochemistry and Molecular Biology and Program in Molecular and Cellular Biology, University of Massachusetts, Amherst, Massachusetts O1 003

~ ~~ ~ ~

We have characterized the expression of two members of a class of Arabidopsis thaliana glycine-rich, putative RNA-binding proteins that we denote Ccrl and Ccr2. Southern blot analysis indicates that Ccrl and Ccr2 are members of a small gene family. Both Ccrl and Ccr2 mRNA levels were influenced by a circadian rhythm that has an unusual phase for plants, with maximal accumulation at 6:OO PM and minimal accumulation at 1000 AM. l h e leve1 of CCRl protein, however, remained relatively constant throughout the cycle. l h e transcript accumulation patterns of the Ccrl and Ccr2 genes differed considerably from conditions that affect the expression of similar genes from maize, sorghum, and carrot. Levels of Ccrl and Ccr2 mRNAs were unchanged in wounded plants, increased at least 4-fold in cold-stressed plants, and decreased 2- to 3-fold in abscisic acid-treated plants. Ccrl transcript levels decreased in response to drought, whereas Ccr2 transcript levels increased under the same conditions. Based on the presence of additional Ccr transcripts in dark-grown plants, we propose that Ccr transcripts may be subjected to a light- or dark-mediated regulation.

GRPs make up one of the major classes of cell wall proteins in plants (Condit and Meagher, 1986; Keller et al., 1988). The expression of plant GRPs can be tissue specific (Quigley et al., 1991; de Oliveira et al., 1993), developmentally regulated (Lei and Wu, 1991; de Oliveira et al., 1993), and responsive to a wide variety of externa1 stimuli, such as virus infection (van Kan et al., 1988; Fang et al., 1991), wounding (Condit and Meagher, 1986; Keller et al., 1988), salicylic acid, water stress (de Oliveira et al., 1990), and light (Kaldenhoff and Richter, 1989).

Severa1 GRPs that lack an amino-terminal signal peptide in their deduced primary structure and, therefore, presumably play a nonstructural role in the cell have been identified in plants. These proteins, which have been described in Arabidopsis thaliana (de Oliveira et al., 1990; van Nocker and Vierstra, 1993), maize embryos (Gómez et al., 1988) and leaves (Didierjean et al., 1992), sorghum seedlings (Cretin and Puigdomenech, 1990), and mature carrot roots (Sturm, 1992), contain two distinct domains: an amino-terminal RNA-binding domain and a Gly-rich carboxy-terminal do-

' Supported by National Science Foundation grants DMB-9004665

* Corresponding author; fax 1-413-545-1812. and DMB-9105890 to A.E.S.

main. Genes encoding nonstructural GRPs are also responsive to environmental stress. mRNA specifying one maize protein accumulates in response to wounding and water stress in leaves and in response to ABA or water stress in embryos (Gómez et al., 1988). Transcripts for the carrot RRM protein accumulate in response to wounding and, to a lesser extent, treatment with ABA (Sturm, 1992). Although the cellular function(s) of these proteins remains unclear, these results suggest a stress-related role for nonstructural GRPs that may involve interaction with specific or nonspecific RNA molecules.

Clues leading to a function for these nonstructural GRPs may come from studies of proteins from other organisms that contain a similar RNA-binding sequence. This RRM (for recent reviews, see Dreyfuss et al., 1988; Bandziulis et al., 1989; Mattaj, 1989; Keene and Query, 1991; Kenan et al., 1991) is composed of 80 amino acids, including two highly conserved sequences, an octomer designated RNPl and a hexamer designated RNP2. Aromatic amino acids within these highly conserved elements have been implicated in direct RNA interactions (Memll et al., 1988). A number of proteins with known or postulated RNA-binding functions and that contain between one and four amino-terminal RRM sequences have been found in organisms ranging from Esch- erichia coli to man (Kim and Baker, 1993).

As with the plant RRM-GRPs, RRM proteins from other organisms are also characterized by distinctive carboxy- terminal domains that can be either acidic, rich in a single amino acid such as Gly or Pro, or contain putative zinc fingers or nucleotide-binding motifs (Swanson and Dreyfuss, 1988; Keene and Query, 1991). Many of these auxiliary domains are proposed to function in protein-protein interactions (Bandziulis et al., 1989), whereas others appear to form separate domains with RNA-binding activity (Kenan et al., 1991). The presence of introns in identical locations within eukaryotic genes that encode RRM proteins (Li et al., 1991; this report) is indicative of an ancient origin for the RRM. Members of this family of RNA-binding proteins have been implicated in a variety of roles in RNA metabolism, including transcription termination (Dombroski and Platt, 1988; Gott-

Abbreviations: Cab, chlorophyll a/b-binding protein; CCR, cold, circadian rhythm, and RNA binding; Col-O, ecotype Columbia wild- type; GRP, glycine-rich protein; RRM, RNA recognition motif; TCV, tumip crinkle virus. -

1015

1016 Carpenter et al. Plant Physiol. Vol. 104, 1994

lieb and Steitz, 1989), mRNA stability (Minvielle-Sebastia et al., 1991), splicing (Query et al., 1989; Scherly et al., 1990; Ge et al., 1991), ribosomal biogenesis (Lapeyre et al., 1987), translational initiation (Milburn et al., 1990), and sex determination (Bell et al., 1988; Amrein et al., 1990).

In an attempt to determine a function for the plant RRM- GRPs, we have initiated a detailed study of two A. thaliana genes that encode RRM-GRPs and appear to be members of a small gene family. Transcripts of the A. thaliana genes accumulated in a11 tissues examined and were influenced by a plant circadian rhythm, with maximal and minimal mRNA accumulation at unusual times in the diumal cycle. Expres- sion of the A. thaliana genes differed from expression pattems of the maize, sorghum, and carrot homologs when exposed to extemal stimuli such as wounding, drought, ABA, and cold. Differential expression pattems in response to the various stress conditions were also evident between the two A. thaliana genes. Preliminary data suggest that Ccr mRNAs accumulate differentially in light-grown versus dark-grown plants.

MATERIALS AND METHODS

Plant Crowth and Treatment Conditions

Arabidopsis thaliana ecotype Columbia wild type (Col-O) was a gift from F. Ausubel (Massachusetts General Hospital, Boston, MA). Plants were grown in controlled environmental chambers with a 16-h light/8-h dark photoperiod (180 pE s-'

unless specified otherwise. Plants with four fully expanded leaves (approximately 2.5 weeks postgermination) were in- dividually inoculated by dipping a glass rod into inoculation buffer containing TCV RNA and rubbing each leaf with a single stroke as previously described (Simon et al., 1992). Mock-treated plants were rubbed in an identical fashion with buffer minus the vira1 RNAs. For ABA treatments, 3-week old plants were sprayed until run-off with 50 m h ABA (mixed isomers) in 0.02% Tween-20 at 6:OO AM. Control plants were sprayed with 0.02% Tween-20 without ABA. Plants were covered with plastic wrap and placed in a growth chamber at 22OC. At 1O:OO AM, roots were excised, and plants were blotted to remove moisture and then stored frozen at -8OOC. For wound treatments, plants were sprinkled with diatoma- ceou earth, and leaves were rubbed vigorously with a glass rod until visible wounding was achieved. Control plants were untreated. Plants were harvested at 1O:OO AM, 22 h later. For drought treatment, plants were grown under standard conditions for 6 d and either watered normally (control) or not watered. Plants subjected to drought conditions had visibly wilted and lost 60% of their fresh weight when compared with rehydrated plants. Plants were harvested at 1O:OO AM. For cold treatment, plants were incubated at 4OC under constant illumination of 20 pE s-' m-' at 1O:OO AM and harvested 24 h later. Control plants were grown under the same light conditions at 22OC. For the cold time course, the same conditions were used except that the plants were subjected to cold stress beginning at 6:OO AM and harvested every 4 h.

m-2. , the light cycle began at 7:OO AM) at a constant 20°C

Screening Arabidopsis Genomic Libraries

An A. thaliana genomic library (ecotype Landsberg) constructed in the vector EMBL4 was a gift from E. EAeyerowitz (Califomia Institute of Technology, Pasadena, CA), and an A. thaliana genomic library (ecotype Col-O in EMBL4) was a gift from C. Town (Case Western Reserve Univeruity, Cleve- land, OH). The Landsberg library was differentially screened (Ausubel et al., 1991) using radiolabeled cDNA prepared from poly(A)+ RNA isolated from leaves of A. thaliana plants 4 d postinoculation with TCV or mock treated with inoculation buffer. The Col-O library was screened using riidiolabeled Ccr2 cDNA. Despite repeated attempts, we were unable to identify phage containing Ccr2 genomic sequences from the Landsberg genomic library.

Construction and Screening of an Arabidopsis cDNA Library

Double-stranded cDNA was prepared using standard methods (Ausubel et al., 1991) from 10 pg of poly(A)+ RNA isolated from A. thaliana ecotype Col-O plants 4 d postinoculation with TCV. The cDNA was treated with EcoRI meth- ylase, ligated to EcoRI linkers, and digested with EcoRI before fractionation in a 1.2% agarose gel. cDNA larger than 900 bp was electroeluted, ligated to predigested X ZapII ECORI arms, and packaged for plating according to the supplier's suggested protocols (Stratagene). An aliquot consisting of 2 x 105 individual recombinant clones was amplified for sub- sequent screening procedures. A 2.8-kb EcoRI-KpnI fragment, from the genomic Ccrl clone, was labeled using random primers (Sen and Murai, 1991) and used as a probe to screen the cDNA library. Two positive clones containing unique inserts were obtained from the 150,000 clones screened.

DNA Sequencing

Genonic and cDNA clones were sequenced by the chain termination method (Sequenase, United States Biochemical) following construction of nested deletions by ExoIII treatment (Ausubel et al., 1991).

DNA lsolation and Cenomic Southern Blots

DNA was isolated from A. thaliana ecotype Col-O as described by Bematzky and Tanksley (1986). DNA (2 rg) was digested with various restriction enzymes and subjected to electrophoresis on a 0.8% agarose gel. The DNA was then transferred to a nylon membrane (Zetaprobe, Bio-Rad) and hybridized to an 883-bp SstI-RsaI fragment derived from the genomic Ccrl clone, or the complete Ccr2 cDNA in:sert, which was labeled using random primers (Sen and Murai, 1991). Hybridization conditions were 5X SSC, 5X Denhardt's solu- tion, 0.2% SDS, 10 m~ NaP04, 0.5 mg mL-' of single- stranded DNA at 68OC ovemight. The blot was then washed for 20 min at 68OC in succession with 2X SSC, 11.1% SDS; l x SSC, 0.1% SDS; and 0.5X SSC, 0.1% SDS.

RNA lsolation and Northern Blots

Total R.NA was isolated from leaves using a LiCl procedure as previously described (Simon et al., 1992). Poly(A)+ RNA

Arabidopsis RNA-Binding Proteins 1017

was isolated following two passages over an oligo(dT)-cellu- lose column (Ausubel et al., 1991). Poly(A)' RNA (8 Pg) or total RNA (4 Pg) was loaded in each lane. RNA was subjected to electrophoresis on standard formaldehyde gels (Sambrook et al., 1989) and then transferred to a NitroPlus 2000 membrane according to the manufacturer's suggested procedure (Micron Separations, Inc., Westboro, MA). Blots were probed and washed as previously described (Simon et al., 1992). Final wash conditions were 0.1X SSPE, 0.1% SDS for 20 min at 5OOC. Autoradiograms were scanned with a two-dimen- sional gel analysis system (Microscan 1000; Technology Re- sources, Nashville, TN), and the data were integrated using software provided by the manufacturer.

Cloning and Expression of CCRl Protein in E. coli

To obtain a full-length reading frame clone of Ccrl (without intron sequence), the 3' end of the cDNA clone extending from the unique BglII site through the poly(A) tail was ligated to the 5' end sequence from the genomic 2.8-kb EcoRI-KpnI fragment, also digested with BglII. The resultant clone served as a substrate in a PCR reaction primed with oligonucleotides that generated a full-length reading frame plus 11 bp of 3' untranslated sequence flanked by BamHI and EcoRI sites, 5' and 3', respectively. The PCR fragment was ligated into the BamHI and EcoRI sites of pGEX-2T (Pharmacia) for expression as a fusion protein with GSH S-transferase. The fusion protein was purified by affinity chromatography using im- mobilized GSH (Smith and Johnson, 1988).

Antibody Production and Western Blot

Six-month-old rabbits were intradermally injected with approximately 1 mg of purified CCRl -GSH S-transferase fusion protein along with Freunds complete adjuvant. After two boosters of 0.5 mg of protein were injected into the rabbits, serum was collected and processed following standard techniques (Ausubel et al., 1991). For western blots, plants were grown and harvested under conditions identical with those used for the northern blot analyses. Proteins were extracted from whole plants as follows: 100 to 300 mg of frozen plant material were ground in liquid nitrogen, followed by further grinding in 30 mM Tris-HC1 (pH 7.2). Debris was removed from the grindate using centrifugation, and the cleared extract was used for analysis. Protein concentration was determined using the Bradford (1976) assay. For the western blot, proteins were subjected to electrophoresis through 12% SDS-polyacrylamide gels and then transferred onto nylon membranes (Nytran) using standard electro- blotting techniques (Ausubel et al., 1991). CCRl protein was detected using the polyclonal antisera described above; a sample of the original antigen was included on each blot as a positive control. The blot was developed using a secondary antibody fused to horseradish peroxidase and stained with the appropriate substrate (4-chloro-1-naphthol, Kierkegaard and Perry).

RESULTS

Cloning and Sequencing of the Ccrl and Ccr2 Genes

This study began as an effort to identify mRNAs that accumulate differentially in response to infection of A. thal-

iana (ecotype Col-O) with TCV. A genomic library prepared from A. thaliana ecotype Landsberg was differentially screened using either cDNA prepared from A. thaliana leaves inoculated with TCV 4 d previously (at the four-expanded leaf stage) or cDNA from plants mock treated with inoculation buffer. One hybridization-positive phage that rescreened successfully was subjected to restriction endonuclease analysis; a 1.1-kb RsaI subfragment within a 2.8-kb EcoRI-KpnI fragment hybridized strongly only to the cDNA from infected plants (data not shown).

The 2.8-kb EcoRI-KpnI fragment was used to screen a cDNA library derived from mRNA isolated from Col-O leaves 4 d after being inoculated with TCV. Of the 150,000 phage screened, two positive cDNA clones containing unique inserts were obtained. One cDNA clone that produced a strong hybridization signal contained a 715-bp inserted sequence and terminated in a poly(A) tract; this sequence was identical with a portion of the genomic sequence contained within the 1.1-kb RsaI fragment, with the exception of a single gap in the cDNA sequence corresponding to a 283-bp A/T-rich segment, flanked by consensus intron-splicing signals (Fig. 1; White et al., 1992). The cDNA sequence was less than full length, containing one open reading frame that extended from one end to 153 bp from the poly(A) tract. The open reading frame terminated with a TAA codon, the most common termination codon in dicots (Cavener and Ray, 1991). Although the 3' untranslated region was rich in thymidine residues, a plant consensus poly(A) addition signal was not present (Joshi, 1987).

Examination of the genomic sequence revealed the pres- ente of an in-frame Met codon 40 nucleotides upstream of the 5' end of the cDNA sequence and 24 nucleotides downstream from a termination codon. There are severa1 lines of evidence that suggest that this is the initiation codon: the sequence immediately preceding the ÃTG, 5'- CTCAAAAAAAAA-3', is very similar to the consensus sequence for translation initiation sites in dicots, 5'- AAAAAAAAAAE-3' (Cavener and Ray, 1991); the sequence 5'-CTTATCA-3', located 63 nucleotides upstream from the presumptive ATG initiation codon, is in agreement with the location and composition of the consensus plant transcription start site (5'-CTCATCA-3'; Joshi, 1987); the sequence 5'-TATAAA-3' is located 18 nucleotides further upstream, in good agreement with the sequence and position of plant TATA boxes (Joshi, 1987). We have designated this gene "Ccrl." A second cDNA clone hybridized weakly to the genomic EcoRI-KpnI fragment derived from Ccrl. This 1075- bp cDNA, designated Ccr2, was used as a probe to isolate the corresponding genomic sequence from a Col-O genomic library. Sequence analysis revealed that the Ccr2 cDNA was identical with the corresponding portion of the Ccr2 genomic sequence and shared approximately 80% sequence similarity with Ccrl cDNA (Fig. IB). This second gene apparently represents a member of the same gene family as Ccrl.

cDNAs corresponding to the Ccrl and Ccr2 genes were recently independently isolated by van Nocker and Vierstra (1993). We noted, as did these authors, the strong similarity to RRM-containing proteins from maize, sorghum, and carrot (see fig. 2 of van Nocker and Vierstra 1993). Comparison of our sequence data for the Ccrl and Ccr2 genes with the


A 90

T 6 k P A C U I T C ~ C A ~ C U I C C C T A ~ C A T G ~ M C C C T f f i T T A T T A C G l C A C A C C T G ~ C ~ ~ ~ C T G C G T ~

A C C T C A A A T C A C A T M G T C T C T C T C ~ A C A ~ 6 k P A C C C T M ~ C T C ~ T G T C T ~ G T T W \ G T A C C G G T G C T ~ G T C G G C

[ 270 GGCC~GCCTGGGCcACCMTMT~MTCTTCI IUGMCGTTCT~CAGTTCGGCMCGTTATCW\~CTM~tctgt t icacgaga

3MI gatcggtctcccggatcgagccgittccgatgattctgitcctcgacggatctgattccgatctgtttctctgtticttgat tcgattac

454 t g t t a c t a t g t t c t c t c t c g t t ~ t t t g t t a c t g t t a c t t a ~ t t t g t c c c i t ~ g g t i c g t t c a t c t t c c t g c t t c t a t ~ a g c t ~ g g a g a t c

%O g a t c g a t t t t t g c t t t a t t t t c a t c g c t t t g t t t t a t i t t c c t t c c i c g a t t g t t t t t g ~ t g ~ t 9 t g t ~ t g ~ t t t t g t t t g t t t ~ c a g A ~

I

L I

M S E V E Y R C F V G

G L A Y A T N D E D L P R T F S Q F G D V I D S K

0 0 CATTMCCACCGCMGAGTGGMMTCMGGGMTTCGGATTCGTCACCTTCMffiMCM~GCcATMGGMTGCMTT~W\MT

I N D R E S G R S R G F G F V T F K D E K A U R D A I E E U

G A A C G G T A A A M G C T C ~ T G G A C G ~ G T ~ l C ~ C C G l ~ C M G G C T C A G T C M ~ ~ T ~ G C G 6 C 6 G T G G C ~ 6 M ~ C G T G G T ~ G N G K E L D G R V I T V N E A P S R G S G G G G G G R G G S

CGGTG6TGGTTACCGCAGCGGCGGTGGTGMTACTCAGGAGGCGGTGGCGGCGMTACTCAGW\GMGGCGGT~TGGTTACW\MG G G G Y R S G G G G G Y S G G G G G G Y S G G G G G G Y E R

9w ACGTAGCGW\GGTTACGW\TCTGGTGMGGCGGTGGTGGCCW\GW\TACGGTGG~GGTGMCGCCGTW\GGMGGTGGCTACGMGGCGG

R S G G Y G S 6 6 6 6 G 6 R G Y G G G G R R E G G G V G G G 990

T W \ T G G T G G A A G T T A C G w 6 G C G G T G G T G G C G G C T G G T ~ T ~ M T A M G ~ G T T T G C G T G C T G C T G C T C T G ~ G G ~ A W \ T ~

110

1111

D G G S Y G G G G G G Y

B O0

T I I U C C M C T C T A A A C C T A ~ T A T C T A A A C C n G G n M T A T C T C A G C C C C C ~ T M C M W \ C n C G T C T A C A T C G n C T A ~ [ I80

~CACT6CTCACTACTCTCACTGTMTCCCTTAMTCTTC~TTCAAA~CMTGGCGTCCGGTMTGTTW\GTATCGGTGCTTC M A S G O V E V R C F

270 G~GMGGTCTAGCAT666CUCTWTWUtAGCTC~W\W\CTGCCTTCGCTCMTACGGCMCGTTA~M~CCMGgtctg t t i c

V G G L A Y A T D D R A L E T A F A Q Y G D V I D S K V C V

Figure 1. Sequence of the A. thaliana Ccrl and CcrZ genes. A, Sequence of t h e Ccrl gene. The coding region is interrupted by a single 283-bp intron. A bracket denotes the 5’ end of the cDNA, and a dot indicates the position of the polyadenylation site in the cDNA. The presumptive TATA box and presumptive transcription start site (based on similarities with pfant consensus sequences) are underlined. The intervening sequence is in lowercase letters. T h e deduced polypeptide sequence is given in one-letter amino acid code under the nucleotide sequence. B, Sequence of the Ccr2 gene. The sequence is presented in the same manner as Ccrl. The intron-like sequence found in our cDNA for CcrZ, but not in the corresponding cDNA isolated by van Nocker and Veirstra (1993), is in lowercase italics, and the amino acids encoded by the intron- like sequence are also in italics.

cDNAs reported by van Nocker and Vierstra indicated five differences in the 5’ and 3’ untranslated regions of Ccrl and two differences in the 5’ and 3’ untranslated regions of Ccr2. No differences were observed in the coding sequences of either gene. These sequence differences probably do not represent differences in source material (our Ccrl gene was isolated from a Landsberg genomic library, and the Ccr2 gene, both of our cDNAs, and the cDNAs of van Nocker and

Vierstra were derived from Col-O) in that our cDNA sequences matched perfectly with the appropriate regions of our genomic sequences and both differed from the corresponding cDNAs of van Nocker and Vierstra. The intervening sequences in Ccrl and Ccr2 are located within the RRM segment.. in the same position as in the genes encoding RRM proteins from tobacco (Li et al., 1991), maize (GOmez et al., 1988), human (Biamonti et al., 1989), and Xenopus (Etzerodt et al., 1988).

A strilting difference between the Ccrl and CcrZ cDNA sequences is the presence of an insertion of 166 bp in the Ccr2 cDNA, corresponding to a portion of the Ccr2 genomic intervening sequence. This 166-bp intervening sequence was not present in the equivalent Ccr2 cDNA reported by van Nocker and Vierstra (1993). Severa1 lines of evideiice suggest that this íntron-derived sequence in our Ccr2 cDNA was not the result of a cloning artifact: (a) the intron segmcnts excised in both Ccr2 cDNAs have correct splice site junction sequences; (b) northern hybrichzation analysis (see below) sug- gests that a second RNA species, slightly larger than the size of fully processed mRNA, is detected using either full-length cDNA or intron-only probes. The presence of tliis “intron- like” sequence in the CcrZ cDNA would have a significant effect on the deduced protein sequence. Protein translated from Ccr2 transcripts corresponding to our cDNA would be severely truncated because of the presence of an in-frame termination codon near the 5‘ end of the intron-derived sequence (Fig. 1B). The resultant protein would have only 48 amino acids with a mo1 wt of 5268. Distinct fornis of RRM- containing proteins have been found in tissues that were generated from alternatively spliced mRNAs (Ge et al., 1991). It is possible that Ccr2 transcripts may also be conditionally processed in a fashion that would generate a truncated and presumably nonfunctional protein product.

Ccrl and Ccr2 Are Members of a Small Cene Family

Genoniic Southern blot analysis was perfor med using either an 883-bp fragment of the Ccrl gene or the 1075-bp Ccr2 cDNA as probes. Arabidopsis DNA was digested with restriction enzymes with recognition sites not found within the Ccrl and Ccr2 genomic segments that were csequenced. The results, presented in Figure 2, revealed one strong hybridization signal in each lane along with as many as six weaker signals (visible in the CcrZ blot only after a prolonged exposure, data not shown). One of the weaker sigrials in each lane corresponded precisely to the strong hybridizing signal of the other cloned gene. These results suggest that Ccrl and Ccr2 are members of a small gene family.

Ccrl Trainscripts Are Not lnduced in Leaves of Virus-lnfected Arabidopsis

Since the Ccrl gene was isolated based on tiifferential hybridization to cDNA from virus-infected plants, RNA gel blot analysis was performed in an attempt to confirm that Ccrl mRIVA accumulated during the response of 14rubidopsis ecotype Col-O to virus infection. Poly(A)+ RNA was prepared from 4-d postinoculation A. thaliana leaves as wdl as from


B

R R

St R

Figure 2. Southern blot analysis of the Ccr/ and Ccr2 genes. A,Cenomic DMA was isolated from leaves of A. thaliana, and 2 jjg perlane were digested with the restriction enzymes indicated. The blotwas probed first with an 883-bp Sstl-Rsal fragment of Ccrl as shownin B, stripped, and then probed with the cDNA insert from Ccr2.Hybridization conditions were identical for both probes with thefinal wash at high stringency (68°C, 0.5X SSC, 0.1 X SDS). Fragmentsizes (in kb), as determined from size marker DNA, are indicatedbetween the blots. B, Diagrams of portions of the Ccr/ and Ccr2genomic clones. Boxes indicate the presumptive transcribed regionsoriented 5' to 3'. The open reading frames are shaded, and thehatched portions indicate intervening sequences. The cross-hatched region in Ccr2 indicates the intron-like sequence from thecDNA. Probes used in A are indicated by lines over the diagrams.B, SamHI; Bl, 6g/ll; D, Oral; E, fcoRI; H, Hindi 11; K, Kpn\; R, Rsal; S,Sstl; St, Styl; X, Xfaal.

leaves of plants treated in an identical fashion with inocula-tion buffer. Based on the location of the putative transcriptioninitiation signal in the Ccrl gene, the expected size of CcrlmRNA is 805 bases plus the poly(A) tail. The Ccrl genomicDNA probe hybridized strongly to an RNA of approximately950 bases and weakly to an RNA of about 3300 bases (Fig.3A). Since the 950-base mRNA species is approximately thesize expected for mRNA transcribed from the Ccrl gene, itwill be referred to as the Ccrl mRNA. To determine whetherthe 3300-base RNA hybridized to sequences throughout theCcrl coding region, a cDNA clone was constructed thatextended from the initiating Met to 11 bp downstream fromthe termination codon. This cDNA was subdivided into twosegments: upstream and downstream from the single Sstl sitewithin the coding sequence (Fig. 2B). Both of the 5' and 3'open reading frame subclones hybridized to the 3300-baseRNA, indicating that this species contains similarities acrossthe entire length of the Ccrl coding sequence (data notshown).

The Ccrl probe did not hybridize differentially to virus-infected and mock-treated RNA preparations (Fig. 3A). Onepossible explanation for this result was that Ccrl transcriptsaccumulated differentially within a short time window (e.g.3.5 or 4.5 d postinoculation) but not precisely when theinfected plants were harvested. Therefore, Arabidopsis plantswere inoculated with TCV or mock treated with buffer, andplants were harvested at 12-h intervals beginning 75 h post-inoculation. The results, presented in Figure 3B, again re-vealed the lack of differential hybridization to RNA frommock-treated and virus-infected plants. However, Ccrl tran-script levels fluctuated depending on the time of day that theplants were harvested. Both infected and mock-treated plantsharvested at 7:00 AM, 87 or 111 h postinoculation, accumu-lated substantially less Ccrl mRNA than plants harvested at7:00 PM, 75 or 99 h postinoculation. Unlike the 950-baseRNA species, the steady-state level of the 3300-base speciesremained constant in these preparations (determined uponoverexposure of the autoradiogram, data not shown).

BMock-treated Virus-infected

3300

^950

Figure 3. RNA gel blot analysis of Ccr 1 mRNA levels in response tovirus infection. A, Poly(A)+ RNA was isolated from A. thaliana 4 dpostinoculation with TCV (Virus-infected) or treated in an identicalfashion with inoculation buffer (Mock-treated). Eight micrograms ofRNA per lane were hybridized with the genomic DMA-derivedprobe described in Figure 2. The approximate size of the twohybridizing species (in bases) was determined by comparison withethidium bromide-stained rRNA and viral RNA species. B, TotalRNA was prepared from plants treated as in A but harvested at thetimes shown above each lane. Four micrograms of RNA per lanewere hybridized first with the Ccrl genomic DNA probe (top) andthen with a pea rDNA probe (Jorgensen et al., 1987; bottom).Arrowheads indicate the positions of the 950- and 3300-base RNAspecies. The 3300-base species was clearly visible with longerexposure.


Figure 4. Ccrl transcript levels varied according to a circadianrhythm. A. thaliana plants were grown for 3 d in a light/dark cycle(top), maintained in continuous light after the first light period(middle), or maintained in continuous dark after the first light period(bottom). Four to six plants were harvested every 4 h. Four micro-grams of total RNA per lane were hybridized to a full-length CcrJcDNA fragment. Uniformity of loading and transfer of the RNA wasconfirmed by reprobing the blots with pea rDNA (Jorgensen et al.,1987). White and black bars below each blot denote periods oflight and dark, respectively. Arrowheads indicate the positions ofthe 950- and 3300-base RNA species. An asterisk (*) denotes theposition of an additional hybridizing species present mainly in plantsgrown under conditions of continuous darkness.

Ccrl and Ccr2 Transcript Levels Are Influenced by a PlantCircadian Rhythm

To examine in greater detail the pattern of Ccrl mRNAaccumulation over time, untreated Ambidopsis plants weregrown under our standard photoperiod of 16 h of light/8 hof dark, at a constant temperature of 20°C, and plants wereharvested at 4-h intervals beginning at 6:00 AM (the lightperiod begins at 7:00 AM). Ccrl mRNA displayed a clearpattern of accumulation, with minimal levels at 10:00 AM andmaximal levels at 6:00 PM (Fig. 4, top). The periodic increaseand decrease in the level of Ccrl transcripts suggested theinvolvement of a circadian rhythm. One of the definingcharacteristics of genes whose patterns of expression areunder control of a circadian clock is the persistence of therhythm in the absence of external cues such as a light/darkcycle (Feldman, 1989). To determine whether the variationobserved for the accumulation of Ccrl mRNA was influencedby a circadian rhythm, plants were grown under a 16-h light/8-h dark photoperiod and then transferred to conditions ofeither continuous light or continuous dark. As shown in

Figure 4 (middle), plants shifted to conditions of continuouslight amassed Ccrl mRNA with a periodicity similar to plantsgrown under a light/dark cycle. A damping of the fluctuationoccurred, beginning in the second cycle in the absence ofenvironmental cues, and the timing of maximal accumulationwas shifted in the third cycle (from 6:00 PM to 10:00 PM).Damping of the fluctuations and shifting of the phase of therhythm was more evident in the dark-grown plants andbegan during the first constant dark cycle (Fig. 4, bottom).Similar fluctuations in transcript accumulation were alsofound for Ccr2 (Fig. 5).

Intron Sequence from Cer2 Hybridizes to a ConstitutivelyExpressed RNA

RNA from dark-grown plants contained additional speciesthat hybridized to the Ccrl and Ccr2 probes. These RNAtranscripts, which migrated slightly slower than the 950-baseRNA species, were also found in several other RNA prepa-rations but not to the same extent as in dark-grown plants. Itwas possible, based on our isolation of a Ccr2 cDNA contain-ing 166 bp of intervening sequence, that the larger RNAspecies represented an alternatively spliced product analo-gous to those found associated with transcripts for otherRRM-containing proteins such as Drosophila sex-lethal (Bellet al., 1988) and transformer (Nagoshi et al., 1988). To testthis, the intervening sequence found in the Ccr2 cDNA wasused to probe the northern blot of RNA from dark-grownplants (Fig. 5, bottom). The result is shown in Figure 6. Theprobe hybridized to an RNA that migrated slower than thefully processed message and co-migrated with the higher-mol wt species detected with the full-length probe. However,the level of the intron-hybridizing species did not fluctuateduring the circadian cycle, and the exposure time required todetect this partially spliced RNA was 3 times longer than forthe dark-abundant RNA. These results suggest that the prom-

Q.

8 8 IpCD

Figure 5. Ccr2 transcript levels also varied according to a circadianrhythm. The blots shown in the middle and bottom panels of Figure4 were reprobed with Ccr2 cDNA. The arrowheads and asterisks (*)are as described in the legend to Figure 4. Hybridization of theCcr2 probe to the 3300-base species was only detectable afterprolonged exposure of the autoradiogram.


Figure 6. Ccr2 intron hybridizes to a larger than full-length RNA.The northern blot of RNA from dark-grown plants used in Figure 5(bottom) was stripped and reprobed with the 170-bp Styl-Rsalfragment from the Ccr2 genomic clone. The second 10:00 PM lanefrom the bottom panel in Figure 5 is reproduced at the left to showthe relative location of the intron-containing band. The hybridiza-tion conditions used were identical with those from Figures 4and 5. The arrows and asterisk are as described in the legend toFigure 4.

inent, larger RNA species detected as increasing in abundancein dark-grown tissue is not due to alternative splicing of theCcrl transcript.

Effect of Circadian Rhythm on the Levels of CCR1 Protein

To determine whether the levels of CCR1 protein alsovaried in a rhythmic fashion, CCR1 protein was purified asa GSH S-transferase fusion protein produced in E. coli, andantibodies to the fusion protein were generated in rabbits.Anti-CCRl antibodies cross-reacted with a single polypeptideof 16 kD, the expected mass of an unprocessed proteinencoded by the Ccrl fully processed transcript (Fig. 7). UnlikeCcrl transcripts, CCR1 protein levels did not fluctuate signif-icantly or reproducibly during a 24-h period.

Effect of Different Stress Conditions on the Levels ofCcrJ and Ccr2 Transcripts

The steady-state levels of transcripts specifying RRM pro-teins from maize and carrot increase during conditions ofwater stress, ABA treatment, or wounding (Gomez et al.,1988; Sturm, 1992). To monitor the accumulation of Ccrl andCcrl mRNAs in response to stress, 3-week-old plants weresubjected to wounding, drought, ABA, or cold. Plants ex-posed to stress conditions were grown under the same light/dark cycle as control plants, and all plants were harvested at10:00 AM, the low point in the circadian cycle. The results arepresented in Figure 8A. Ccrl and Ccrl mRNAs did notaccumulate differentially in response to wounding. It shouldbe noted that harvesting the plants at the low point in thecircadian cycle might preclude observing an induction if thecircadian regulation overrides a stress-induction signal. How-ever, we did observe responses to other stresses at 10:00 AM(see below), suggesting that the circadian regulation of theCcr genes does not prevent responding to other stimuli. CcrlmRNA levels decreased in plants treated with ABA or sub-jected to water stress, whereas Ccrl transcript levels increasedin water-stressed plants. Plants exposed to cold stress showeda marked increase in the amounts of both Ccrl and Ccr2transcripts. As a control, the blot in Figure 8A was alsoprobed with pHH29, which had previously been shown to

hybridize to an RNA species that accumulated in response todrought, ABA, and cold (Hajela et al., 1990); these resultswere repeated under our stress conditions.

To assess the kinetics of Ccrl and Ccrl mRNA accumulationover time due to cold treatment, plants were placed at 4°Cat 6:00 AM and harvested at 4-h intervals (Fig. 8B). Asexpected, Ccrl and Ccrl mRNAs from control plants accu-mulated higher levels over time because of the circadianfluctuations described above. However, Ccrl and Ccrl tran-scripts in leaves from plants incubated at 4°C for 12 and 16h were more abundant than in control plants. After wenormalized for the amount of RNA loaded and subtractedthe signal due to circadian effects observed in the controls,transcript levels for Ccrl and Ccrl increased approximately4-fold in plants incubated for 16 h at 4°C. The timing of theincreased accumulation of Ccrl and Ccrl mRNAs was lessrapid when compared with mRNA hybridizing to pHH29,which began accruing after 4 h of cold treatment (Fig. 8B).The RNA levels for pHH29 may be regulated by a circadianrhythm (Fig. 8B), but confirmation will require further analy-sis. Computer analysis of the upstream sequences from Ccrland Ccr2 did not yield any strong similarities with consensussequences of other promoters that are known to be inducible(as defined by Cattivelli and Bartels, 1992; Olsen et al., 1992;Schindler et al., 1992).

B

.8 8 8 8 •§ID T- CM <D ,- CM 5

Figure 7. Western blot analysis of CCR1 protein levels. Proteinextracts were made from plants harvested at 4-h intervals (times aregiven above the lanes). Protein (20 Mg) was subjected to electro-phoresis on a 12% SDS-polyacrylamide gel; one set of two duplicatesamples was stained with Coomassie blue to determine loadinguniformity (A), and the second set was used to prepare a westernblot (B). Numbers on the sides denote mol wts of protein markers(X10~3). CCR1 protein was detected using rabbit polyclonal antiseragenerated using recombinant CCR1 fused to CSH S-transferase. Asample of the original antigen was included as a positive control(fusion; expected mol wt of 42 X 103).


Ccr1

Ccr2

pHH29

rDNA

B 22°C 4°C

Ccr1

Ccr2

PHH29

rDNA

Figure 8. Effects of wounding, drought, ABA, and cold on Ccrl andCcr2 transcript levels. Stress treatments are described in "Materialsand Methods." A, A single blot was probed in succession with Ccrlgenomic DMA, Ccr2 cDNA, pHH29 (Hajela et al., 1990), and pearDNA. pHH29 is a cDNA clone from A. thaliana whose transcriptlevels have previously been shown to increase following treatmentwith cold, ABA, or drought (Hajela et al., 1990). Plant treatmentswith ABA were repeated three times with very similar results. B,Time course for the accumulation of Ccrl and Ccr2 transcripts inresponse to cold. Plants were grown under constant light (beginningat 6:00 AM), and leaves were harvested at the times indicated.

Accumulation of Ccrf and Ccr2 Transcripts in DifferentArabidopsis Tissues

To determine whether Ccrl and Ccrl are expressed in atissue-specific manner, RNA was extracted from Arabidopsisleaves, flowers, siliques, and stems at both 10:00 AM and 6:00PM (low and high transcript accumulation points in the cir-cadian cycle). The results, presented in Figure 9, indicate thatCcrl and Ccrl are expressed at comparable levels in all planttissues tested, and both are still regulated by the circadianrhythm. Curiously, the transcript that migrated slightlyslower than the 950-base transcript was not detected in RNAisolated from silique tissue probed with either Ccrl or Ccr2.Rather, a new larger transcript that hybridized to the Ccrl

probe was found in silique RNA preparations at both 10:00AM and 6:00 PM. This transcript could represent an additionalfamily member or an alternatively processed form of CcrlmRNA.

DISCUSSION

van Nocker and Viersrra (1993) recently reported the se-quences of two A. thaliana cDNAs that contained RRMs. Wereport here the independent isolation and characterization oftwo very similar, and possibly identical, A. thaliana genesthat encode the cDNAs. Based on the regulation of expressionof these genes, we have named them Ccrl and Ccr2 (Cold,Circadian, RNA binding). Ccrl and Ccr2 mRNA levels exhib-ited cyclic variation that was maintained under constantenvironmental conditions. This pattern of expression indi-cated that the Ccrl and Ccr2 genes are under control of anendogenous circadian clock. A number of processes in plantsare affected by a biological timer, including photosynthesis,leaf positioning, and sap exudation (reviewed by Feldman,1989). Cab, encoding the major light-harvesting Chl-bindingprotein of the chloroplast, is expressed at maximal levels soonafter the onset of light in wheat (Nagy et al., 1988), tomato

d>r 0 -5

-5 | fS g 55_ _

106 10 6 106 106 106

Iprobe: Ccrl

probe: Ccr2

probe: rDNA

Figure 9. Ccrl and Ccr2 transcript levels in different tissues. RNAwas isolated from various tissues (listed above the lanes; whole =whole plants). Tissues were harvested at the low (10:00 AM) andhigh points (6:00 PM) in the circadian cycle of Ccrl and Ccr2 (timeslisted above each lane). A northern blot was produced, probed withthe Ccrl genomic probe (see Fig. 2B), stripped, probed with thecDNA clone for Ccr2, stripped, and reprobed with the pea rDNAgene. Arrow and asterisk symbols are defined in the legend toFigure 4. The open asterisks denote a new Ccr7-hybridizing speciesfound only in RNA prepared from siliques.


(Piechulla, 1988), and Arabidopsis (Millar and Kay, 1991). The pattern of accumulation of Ccrl and Ccr2 mRNAs, with minimal levels occumng 3 h following illumination and maximal levels 11 h following illumination, is distinctly different from the expression pattern of Cab. The circadian rhythm for the Ccr genes is also different from that observed for the nitrate reductase genes in Arabidopsis (Cheng et al., 1991) in which the RNA levels peaked in the early morning and decreased during the day to reach a low point at the end of the light period. It is interesting that the cyclic expression pattern of the Ccrl and Ccr2 genes is very similar to the pattern of expression of the maize catalase 3 gene (Redin- baugh et al., 1990).

Ccrl and Ccr2 transcript levels also appeared to be influenced by cold treatment. Plants incubated at 4OC for 24 h beginning at 1O:OO AM amassed significantly more Ccrl and Ccr2 transcripts than plants grown at 22OC (Fig. 8A). Caution must be used, however, when interpreting these results. Martino-Catt and Ort (1992) found that the circadian rhythm responsible for fluctuations in expression of Cab was suspended during cold treatment in tomato, a highly chill- sensitive plant. Furthermore, the turnover of Cab mRNA was suspended during the cold treatment. However, changes in environmental conditions have been shown to transiently affect circadian regulation in the short term, followed by a resumption of normal cycling after prolonged exposure (Pit- tenrigh, 1993, and refs. therein). We are in the process of determining Ccr RNA levels in cold-treated plants after sev- era1 days of exposure to the reduced temperature. It should be noted that the RNA levels for Ccrl and Ccr2 increased during the cold treatment above the levels for the corresponding times in the control tissues. These results suggest either a cold responsiveness in the control of transcription or a cold- sensitive RNA degradation component of the circadian rhythm regulation.

Southern blot analysis revealed that Ccrl and Ccr2 are members of a small gene family. The 3300-base RNA species that hybridized to fragments from both 5’ and 3’ end-coding sequences derived from Ccrl did not accumulate differentially in response to cold or the circadian cycle and may represent an additional family member. In dark-grown plants and in the circadian cycle when maximal levels of Ccr transcripts are produced, an additional RNA species was observed that hybridized strongly to both Ccrl and Ccr2 probes and migrated slightly slower than the 950-base transcript. A second RNA species, which co-migrated with the dark-abundant RNA species, was detected when the northern blot of RNA from dark-grown plants was probed with a fragment of the Ccr2 intron that was found in the Ccr2 cDNA. However, based on the low abundance of this intron-containing sequence and its lack of circadian regulation, this transcript is unlikely to be the same as the larger transcript detected with the full-length Ccr2 probe. Other explanations for the identity of the dark-abundant transcript include another member of the Ccr gene family or a different transcription start site. We are currently working to identify the additional cycling RNA species that accumulates in dark-grown plants.

The Ccrl gene was isolated from a genomic library by differential screening using cDNAs generated from RNA isolated from A. thaliana that had been either inoculated 4 d

previously with TCV or mock treated with buffer. RNA gel blot analysis using RNA preparations differing from those used in the initial library screenings revealed that Ccrl mRNA did not accumulate in response to virus infection as originally thought. What then was the difference between the original mock-treated and virus-infected plants? One possible explanation was the way in which the plants used to prepare RNA for the first screen were harvested. The severa1 thousand mock-treated plants required for the generation of the cDNA probes used in the initial differential screening of the genomic library were harvested in the late moming and an equal number of virus-infected plants were harvested in the early afternoon. As shown in Figure 4, levels of Ccrl transcripts fluctuated according to a plant circadian rhythm with minimal amounts present at 1O:OO AM and maximal amounts at 6:OO PM. Plants harvested in the late moming would, therefore, contain less Ccrl mRNA than plants harvested in the afternoon.

Although this explanation seems promising, it should be noted that a second genomic clone isolated in the same screen did not accumulate in response to the virus or the plant circadian rhythm. This clone has been determined to encode a nove1 Gly-rich protein with sequence similarity to group 2 late embryogenesis abundant proteins (C.D. Carpenter and A.E. Simon, unpublished data). Therefore, it is possible that the two populations of mRNA from virus-infected and mock- treated plants differed in another, yet undefined, parameter.

The cellular role(s) of this class of plant Gly-rich RRM proteins remains a mystery. Many of the RRM proteins from other organisms have a binding affinity for specific target sequences such as the 70K U1 snRNP protein, which binds with high affinity to the 5’ half of U1 RNA (Query et al., 1989). Other RRM proteins, such as the poly(A)-binding protein, bind to mRNAs in general (Adam et al., 1986). One possible role for CCRl may be to stabilize mRNAs, either specific species or, in general, during conditions of cold or other, as yet undetermined, environmental stresses. We have recently demonstrated that CCRl can indeed bind RNA (C.D. Carpenter and A.E. Simon, unpublished data) and are in the process of determining whether CCRl preferentially binds to specific RNAs and/or sequences as a prerequisite to ascer- taining a functional role for the protein. It is interesting that the increase in mRNA levels for other A. thaliana cold- inducible proteins involves a posttranscriptional mechanism (Hajela et al., 1990) that could be mediated by binding to a protein such as CCR1. A second possible role for CCRl may be in activating mRNAs that are translationally suppressed because of secondary structure or other constraints. This speculation is based on early reports that RRM-containing proteins can function as helix-unwinding proteins (Hemck and Alberts, 1976). Work is currently under way to determine the RNA-binding capabilities of CCRl and to distinguish between these and other possible functions for the protein.

ACKNOWLEDCMENTS

We thank Susan Martino-Catt for helpful discussions, Michael F. Thomashow for providing the plasmid pHH29 and for useful discussions, and Arnd Sturm for providing a preprint of his manuscript.


Received June 25, 1993; accepted November 23, 1993. Copyright Clearance Center: 0032-0889/94/104/1015/11.

LITERATURE ClTED

Adam SA, Nakagawa TY, Swanson MS, Woodruff T, Dreyfuss G (1986) mRNA polyadenylate-binding protein: gene isolation, sequencing and identification of a ribonucleoprotein consensus sequence. Mo1 Cell Biol6 2932-2943

Amrein H, Maniatis T, Nothiger R (1990) Altematively spliced transcripts of the sex-detennining gene tra-2 of Drosophila encode functional proteins of different size. EMBO J 9 3619-3629

Ausubel FM, Brent R, Kingston RE, Moore DD, Seidman JG, Smith JA, Struhl K (1991) Current Protocols in Molecular Biology. Greene Publishing Associates/Wiley Interscience, New York

Bandziulis RJ, Swanson MS, Dreyfuss G (1989) RNA-binding proteins as developmental regulators. Genes Dev 3 431-437

Bradford M (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein- dye binding. Ana1 Biochem 7 2 248-254

Bell LR, Maine EM, Schedl P, Cline TW (1988) Sex-lethal, a Drosophila sex determination switch gene, exhibits sex-specific RNA splicing and sequence similarity t o RNA binding prbteins. Cell55 1037-1046

Bernatzky R, Tanksley SD (1986) Genetics of actin-related sequences in tomato. Theor Appl Genet 72: 314-321

Biamonti G, Buvoli M, Bassi MT, Morandi C, Cobianchi F, Riva S (1989) Isolation of an active gene encoding human hnRNP protein Al . Evidence for altemative splicing. J Mo1 Biol 207:

Cattivelli L, Bartels D (1992) Biochemishy and molecular biology of cold-inducible enzymes and proteins in higher plants. In JL Wray, ed, Inducible Plant Proteins: Their Biochemistry and Molec- ular Biology. Cambridge University Press, New York, pp 267-288

Cavener DR, Ray SC (1991) Eukaryotic start and stop translation sites. Nucleic Acids Res 1 9 3185-3192

Cheng C, Acedo GN, Dewdney J, Goodman HM, Conkling MA (1991) Differential expression of the two Arabidopsis nitrate reductase genes. Plant Physiol96 275-279

Condit CM, Meagher RB (1986) A gene encoding a novel glycine- rich structural protein of petunia. Nature 323 178-181

Cretin C, Puigdomenech P (1990) Glycine-rich RNA-binding proteins from Sorghum vulgure. Plant Mo1 Bioll5 783-785

de Oliveira DE, Franco LO, Simoens C, Seurinck J, Coppieters J, Botterman J, Van Montagu M (1993) Inflorescence-specific genes from Arabidopsis tholiana encoding glycine-rich proteins. Plant J 3:

de Oliveira DE, Seurinck J, Inze D, Van Montagu M, Botterman J (1990) Differential expression of five Arabidopsis genes encoding glycine-rich proteins. Plant Cell2 427-436

Didierjean L, Frendo P, Burkard G (1992) Stress responses in maize: sequence analysis of cDNAs encoding glycine-rich proteins. Plant Mo1 Biol 18: 847-849

Dombroski AJ, Platt T (1988) Structure of the rho factor: an RNA- binding domain and a separate region with strong similarity to proven ATP-binding domains. Proc Natl Acad Sci USA 85:

Dreyfuss G, Swanson MS, Pinol-Roma S (1988) Heterogeneous nuclear ribonucleoprotein particles and the pathway of mRNA formation. Trends Biochem Sci 1 3 86-91

Etzerodt M, Vignali R, Ciliberto G, Scherly D, Mattaj W, Philipson L (1988) Structure and expression of a Xenopus gene encoding an snRNP protein (U1 70K). EMBO J 7: 4311-4321

Fang RX, Pang Z, Gao DM, Mang KQ, Chua NH (1991) cDNA sequence of a virus-inducible, glycine-rich protein gene from rice. Plant Mo1 Biol 17: 1255-1257

Feldman JF (1989) Circadian rhythms. In KC Smith, ed, The Science of Photobiology. Plenum Press, New York, pp 193-213

Ge H, Zuo P, Manley JL (1991) Primary structure of the human splicing factor ASF reveals similarities with Drosophila regulators. Cell66 373-382

GÓmez J, Sanchez-Martinez D, Stiefel V, Rigau J, Puigdomenech

491-503

495-507

2538-2542

P, Pagecr M (1988) A gene induced by the plant hormone abscisic acid in response to water stress encodes a glycine-rich protein. Nature 334 262-264

Gottlieb E, Steitz JA (1989) Function of the mammalian La protein: evidence for its action in transcription termination by RNA polym- erase 111. EMBO J 8: 851-861

Hajela RK, Horvath DP, Gilmour SJ, Thomashow MF (1990) Molecular cloning and expression of cor (cold-regulated) genes in Arubidopsis fhuliunu. Plant Physiol 93: 1246-1252

Herrick G, Alberts B (1976) Nucleic acid helix-coil transitions mediated by hehx unwinding proteins from calf thymus. .I Biol chem 251: 2133-2141

Jorgensen RA, Cuellar RE, Thompson WF, KavanaghL TA (1987) Structure and variation in ribosomal RNA genes of peii. Plant Mo1 Biol8 3-12

Joshi CP (1987) An inspection of the domain behveen putative TATA box and translation start site in 79 plant genes. Nucleic Acids Res 1 5 6643-6653

Ka1denho:ff R, Richter G (1989) Sequence of a cDNA for a novel light-induced glycine-rich protein. Nucleic Acids Res 1.7: 2853

Keene JDp Query CC (1991) Nuclear RNA-binding proteins. Prog Nucleic Acid Res 41: 179-202

Keller B, Schmid J, Lamb CJ (1988) Glycine-rich cell-wall proteins in bean: gene structure and association of the protein with the vascular system. EMBO J 7: 3625-3633

Kenan DJ,, Query CC, Keene JD (1991) RNA recognition: towards identifying determinants of speaficity. Trends Biochem Sci 16:

Kim Y, Baker BS (1993) Isolation of RRM-type RNA-bintiingprotein genes and the analysis of their relatedness by using a numerical approach. Mo1 Cell Bioll3 174-183

Lapeyre B, Bourbon H, Amalric F (1987) Nucleolin, the major nucleolar protein of growing eukaryotic cells: an unusual protein structure revealed by the nucleotide sequence. Proc Niltl Acad Sci

Lei M, Wii R (1991) A novel glycine-rich cell wall protein gene in rice. Plant Mo1 Biol 1 6 187-198

Li Y, Ye L, Sugita M, Sugiura M (1991) Tobacco nuclear gene for the 31 Itd chloroplast ribonucleoprotein: genomic organization, sequence analysis and expression. Nucleic Acids Res 1 9

Martino-C:att S, Ort DR (1992) Low temperature interrupts drcadian regulation of transcriptional activity in chilling-sensitive plants. Proc Natl Acad Sci USA 8 9 3731-3735

Mattaj IW (1989) A binding consensus: RNA-protein interactions in splicing snRNPs, and sex. Cell57: 1-3

Merrill BM, Stone KL, Cobianchi F, Wilson SH, Wiilliams KR (1988) Phenylalanines that are conserved among severa1 RNA- binding proteins form part of a nucleic acid-binding pocket in the A1 heterogeneous nuclear ribonucleoprotein. J Biol Chem 263

Milburn SC, Hershey JWB, Davies MV, Kelleher K, Kaufman RJ (1990) Cloning and expression of eukaryotic initiation factor 4B cDNA: sequence determination identifies a common RNA recognition motif. EMBO J 9 2783-2790

Millar AJ, Kay SA (1991) Circadian control of cab gene transcription and mRNA accumulation in Arabidopsis. Plant Cel l3 541-550

Minvielle-Sebastia L, Winsor B, Bonneaud N, Lacroute F (1991) Mutations in the yeast RNA14 and RNA15 genes result in an abnormal mRNA decay rate; sequence analysis reveals an RNA- binding domain in the RNA15 protein. Mo1 Cell Biol 11:

Nagoshi R:N, McKeown M, Burtis KC, Belote JM, Baker BS (1988) The control of altemative splicing at genes regulating sexual dif- ferentiation in D. melanogaster. Cell 5 3 229-236

Nagy F, K:ay SA, Chua NH (1988) A drcadian cloclc regulates transcription of the wheat Cab-1 gene. Genes Dev 2 376-382

Olsen FL, Eikriver K, Miiller-Uri F, Raikhel NV, Rogers JC, Mundy J (1992) 14BA and GA-responsive gene expression. In JL Wray, ed, Inducible! Plant Proteins: Their Biochemistry and Molc!cular Biol- ogy. Cambridge University Press, New York, pp 139-154

Piechulla B (1988) Plastid and nuclear mRNA fluctuations in tomato

214-220

USA 84: 1472-1476

2987-2991

3307-3313

3075-3087


leaves-diumal and circadian rhythms during extended dark and light periods. Plant Mo1 Biol 11: 345-353

Pittenrigh CS (1993) Temporal organization: reflections of a darwin- ian clock-watcher. Annu Rev Physiol 5 5 17-54

Query CC, Bentley RC, Keene JD (1989) A common RNA recognition motif identified within a defined U1 RNA binding domain of the 70K U1 snRNp protein. Cell57: 89-101

Quigley F, Villiot ML, Mache R (1991) Nucleotide sequence and expression of a nove1 glycine-rich protein gene from Arabidopsis thaliana. Plant Mo1 Biol17: 949-952

Redinbaugh MG, Sabre M, Scandalios JG (1990) Expression of the maize Cat3 catalase gene is under the influence of a circadian rhythm. Proc Natl Acad Sci USA 87: 6853-6857

Sambrook J, Fritsch EF, Maniatis T (1989) Molecular Cloning. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY

Scherly D, Boelens W, Dathan NA, van Venrooij WJ, Mattaj IW (1990) Major determinants of the specificity of interaction between small nuclear ribonucleoproteins U1A and U2B’ and their cognate RNAs. Nature 345 502-506

Schindler U, Menkens AE, Cashmore AR (1992) GBF-1, GBF-2 and GBF-3: three Arabidopsis b-Zip proteins that interact with the light-regulated rbcS-IA promoter. Zn JL Wray, ed, Inducible Plant Proteins: Their Biochemistry and Molecular Biology. Cambridge University Press, New York, pp 289-304

Sen P, Murai N (1991) Oligolabeling DNA probes to high specific activity with Sequenase. Plant Mo1 Biol Rep 9 127-130

Simon AE, Li XH, Lew JE, Stange RE, Zhang C, Polacco M, Carpenter CD (1992) Susceptibility and resistance of Arabidopsis thalinna to tumip crinkle virus. Mo1 Plant-Microbe Interact 5

Smith DB, Johnson KS (1988) Single-step purification of polypep- tides expressed in Escherichia coli as fusions with glutathione S- transferase. Gene 67: 31-40

Sturm A (1992) A wound-inducible glycine-rich protein from Daucus cnrota with homology to single-stranded nucleic aad-binding proteins. Plant Physiol 9 9 1689-1692

Swanson MS, Dreyfuss G (1988) RNA binding specificity of hnRNP proteins: a subset bind to the 3’ end of introns. EMBO J 7:

van Kan JAL, Cornelissen BJC, Bol JF (1988) A virus-inducible tobacco gene encoding a glycine-rich protein shares putative reg- ulatory elements with the ribulose bisphosphate carboxylase small subunit gene. Mo1 Plant Microbe Interact 1: 107-112

van Nocker S, Vierstra RD (1993) Two cDNAs from Arabidopsis thaliana encode putative RNA binding proteins containing glycine- rich domains. Plant Mo1 Biol21: 695-699

White O, Soderlund C, Shanmugan P, Fields C (1992) Information contents and dinucleotide compositions of plant intron sequences vary with evolutionary origin. Plant Mo1 Biol9 1057-1064

496-503

3519-3529

G e n es E n cod i n g G I y ci n e - Ri c h A rabidopsis thaliana

Documents